New reporting suggests that a recent decline in ChatGPT-attributed website traffic may be caused by a measurement problem rather than a real reduction in visits from ChatGPT. According to the source, analysts linked the drop to ChatGPT links opening without tracking tags, which can make visits harder to classify correctly in analytics platforms.

What happened

The reported issue concerns traffic attribution — the process analytics tools use to identify where website visitors came from. If a visit arrives without the expected referral information or tracking parameters, it may be grouped into other channels such as direct traffic, unassigned traffic, or another default category depending on the analytics setup.

Based on the source, the apparent fall in ChatGPT referral traffic was observed in several datasets, but the explanation presented is that some ChatGPT-generated links may now open in a way that does not consistently pass tracking data. That means the visible decline in dashboards may not reflect an equivalent decline in actual user visits.

Why it matters for European businesses

For European companies increasingly monitoring AI search and generative AI discovery, attribution quality matters. Many marketing and e-commerce teams are trying to understand whether platforms such as ChatGPT are sending meaningful traffic, generating assisted conversions, or influencing brand discovery earlier in the buying journey.

If tracking is incomplete, businesses may draw the wrong conclusions from their analytics. A company could underestimate the value of AI-driven referral traffic, misjudge SEO performance, or shift budget away from channels that are still contributing visits and leads.

This is especially relevant for SMEs and marketing teams that rely on standard analytics dashboards rather than custom attribution models. If AI-originated visits are being misclassified, monthly performance reports may understate the role of ChatGPT and similar platforms in website discovery.

Who may be affected

The issue may be most relevant for:

• Marketing teams tracking referral growth from AI platforms
• SEO teams measuring visibility in AI-assisted discovery environments
• E-commerce businesses analysing product-page visits and assisted conversions
• Digital agencies reporting traffic sources to clients
• IT and analytics teams responsible for attribution, tagging and reporting accuracy

Businesses with a growing share of traffic from ChatGPT or other AI interfaces may see the largest reporting distortion. Companies that compare month-on-month traffic by source should be cautious if they notice unexplained declines limited to AI referral categories.

What companies should consider

• Do not assume a traffic drop is real without validation. Compare analytics reports with landing-page trends, server-side logs and conversion data where available.
• Review channel classification. Check whether visits that previously appeared as ChatGPT referrals are now appearing as direct, unassigned or other source categories.
• Monitor AI traffic with multiple signals. In addition to standard referral reports, track landing pages, assisted conversions, branded search lift and on-site behaviour from suspected AI-originated sessions.
• Document reporting limitations. Agencies and in-house teams should explain that attribution for AI platforms can change due to technical behaviour outside the company’s control.
• Avoid premature budget or strategy changes. If AI referral numbers fall sharply but other engagement indicators remain stable, it may be better to investigate measurement before changing SEO or content priorities.

For European businesses, the main takeaway is practical: as AI platforms become part of the customer journey, measurement reliability becomes as important as traffic volume itself. Reported changes in AI referral traffic should be interpreted carefully, especially when the underlying issue may be technical attribution rather than user demand.